Pub Date : 2026-02-05DOI: 10.1038/s41562-025-02324-0
Maria K Eckstein, Christopher Summerfield, Nathaniel D Daw, Kevin J Miller
A long-standing challenge for psychology and neuroscience is to understand the transformations by which past experiences shape future behaviour. Reward-guided learning is typically modelled using simple reinforcement learning (RL) algorithms. In RL, a handful of incrementally updated internal variables both summarize past rewards and drive future choice. Here we describe work that questions the assumptions of many RL models. We adopt a hybrid modelling approach that integrates artificial neural networks into interpretable cognitive architectures, estimating a maximally general form for each algorithmic component and systematically evaluating its necessity and sufficiency. Applying this method to a large dataset of human reward-learning behaviour, we show that successful models require independent and flexible memory variables that can track rich representations of the past. Using a modelling approach that combines predictive accuracy and interpretability, these results call into question an entire class of popular RL models based on incremental updating of scalar reward predictions.
{"title":"Hybrid neural-cognitive models reveal how memory shapes human reward learning.","authors":"Maria K Eckstein, Christopher Summerfield, Nathaniel D Daw, Kevin J Miller","doi":"10.1038/s41562-025-02324-0","DOIUrl":"https://doi.org/10.1038/s41562-025-02324-0","url":null,"abstract":"<p><p>A long-standing challenge for psychology and neuroscience is to understand the transformations by which past experiences shape future behaviour. Reward-guided learning is typically modelled using simple reinforcement learning (RL) algorithms. In RL, a handful of incrementally updated internal variables both summarize past rewards and drive future choice. Here we describe work that questions the assumptions of many RL models. We adopt a hybrid modelling approach that integrates artificial neural networks into interpretable cognitive architectures, estimating a maximally general form for each algorithmic component and systematically evaluating its necessity and sufficiency. Applying this method to a large dataset of human reward-learning behaviour, we show that successful models require independent and flexible memory variables that can track rich representations of the past. Using a modelling approach that combines predictive accuracy and interpretability, these results call into question an entire class of popular RL models based on incremental updating of scalar reward predictions.</p>","PeriodicalId":19074,"journal":{"name":"Nature Human Behaviour","volume":" ","pages":""},"PeriodicalIF":15.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146125809","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-05DOI: 10.1038/s41562-026-02405-8
Ilias G Pechlivanidis, Spyros Afentoulidis, Giuliano Di Baldassarre, Florian Pappenberger, Peter Salamon, Stefan Uhlenbrook
{"title":"How personalized disaster warnings can save lives.","authors":"Ilias G Pechlivanidis, Spyros Afentoulidis, Giuliano Di Baldassarre, Florian Pappenberger, Peter Salamon, Stefan Uhlenbrook","doi":"10.1038/s41562-026-02405-8","DOIUrl":"https://doi.org/10.1038/s41562-026-02405-8","url":null,"abstract":"","PeriodicalId":19074,"journal":{"name":"Nature Human Behaviour","volume":" ","pages":""},"PeriodicalIF":15.9,"publicationDate":"2026-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146125785","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-02DOI: 10.1038/s41562-025-02397-x
{"title":"Generative AI predicts personality traits on the basis of open-ended narratives.","authors":"","doi":"10.1038/s41562-025-02397-x","DOIUrl":"https://doi.org/10.1038/s41562-025-02397-x","url":null,"abstract":"","PeriodicalId":19074,"journal":{"name":"Nature Human Behaviour","volume":" ","pages":""},"PeriodicalIF":15.9,"publicationDate":"2026-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146106338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-02DOI: 10.1038/s41562-026-02404-9
Fanli Jia
{"title":"The tension between big team science and colonial power dynamics.","authors":"Fanli Jia","doi":"10.1038/s41562-026-02404-9","DOIUrl":"https://doi.org/10.1038/s41562-026-02404-9","url":null,"abstract":"","PeriodicalId":19074,"journal":{"name":"Nature Human Behaviour","volume":" ","pages":""},"PeriodicalIF":15.9,"publicationDate":"2026-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146106378","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-02DOI: 10.1038/s41562-026-02401-y
John F Helliwell, Richard Layard, Jeffrey D Sachs, Jan-Emmanuel De Neve, Lara B Aknin, Shun Wang
{"title":"Why single-item measures of wellbeing are best.","authors":"John F Helliwell, Richard Layard, Jeffrey D Sachs, Jan-Emmanuel De Neve, Lara B Aknin, Shun Wang","doi":"10.1038/s41562-026-02401-y","DOIUrl":"https://doi.org/10.1038/s41562-026-02401-y","url":null,"abstract":"","PeriodicalId":19074,"journal":{"name":"Nature Human Behaviour","volume":" ","pages":""},"PeriodicalIF":15.9,"publicationDate":"2026-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146106387","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Political action is now crucial for US scientists.","authors":"Tatiane Russo-Tait, Summer Blanco, Eduardo Bonilla-Silva","doi":"10.1038/s41562-026-02406-7","DOIUrl":"https://doi.org/10.1038/s41562-026-02406-7","url":null,"abstract":"","PeriodicalId":19074,"journal":{"name":"Nature Human Behaviour","volume":" ","pages":""},"PeriodicalIF":15.9,"publicationDate":"2026-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146106380","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-02-02DOI: 10.1038/s41562-026-02419-2
Anatolia Batruch, Nicolas Sommet, Frédérique Autin
{"title":"Author Correction: Advancing the psychology of social class with large-scale replications in four countries.","authors":"Anatolia Batruch, Nicolas Sommet, Frédérique Autin","doi":"10.1038/s41562-026-02419-2","DOIUrl":"https://doi.org/10.1038/s41562-026-02419-2","url":null,"abstract":"","PeriodicalId":19074,"journal":{"name":"Nature Human Behaviour","volume":" ","pages":""},"PeriodicalIF":15.9,"publicationDate":"2026-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146106355","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-30DOI: 10.1038/s41562-025-02389-x
Aidan G C Wright,Whitney R Ringwald,Colin E Vize,Johannes C Eichstaedt,Mike Angstadt,Aman Taxali,Chandra Sripada
Contemporary personality assessment relies heavily on psychometric scales, which offer efficiency but risk oversimplifying the rich and contextual nature of personality. Recognizing these limitations, this study explores the use of commercially available generative large language models (LLMs), such as ChatGPT, Claude and so on, to assess personality traits from open-ended qualitative narratives. Across two distinct samples and methodologies (spontaneous streams of thought and daily video diaries), we used seven commercial, generative LLMs to score Big-Five personality traits, achieving convergence with self-report measures comparable to or exceeding established benchmarks (for example, self-other agreement, ecological momentary assessment, and bespoke machine learning models). Although results differed across different LLMs, we found that using the average LLM score across models provided the strongest agreement with self-report. Further, LLM-generated trait scores also demonstrated predictive validity regarding daily behaviours and mental health outcomes. This LLM-based approach achieved quantitative rigour based on qualitative data and is easily accessible without specialized training. Importantly, our findings also reaffirm that personality is expressed ubiquitously, in that it is carried in the stream of our thoughts and is woven into the fabric of our daily lives. These results encourage broader adoption of generative LLMs for psychological assessment and-given the new generation of tools-stress the value of idiographic narratives as reliable sources of psychological insight.
{"title":"Assessing personality using zero-shot generative AI scoring of brief open-ended text.","authors":"Aidan G C Wright,Whitney R Ringwald,Colin E Vize,Johannes C Eichstaedt,Mike Angstadt,Aman Taxali,Chandra Sripada","doi":"10.1038/s41562-025-02389-x","DOIUrl":"https://doi.org/10.1038/s41562-025-02389-x","url":null,"abstract":"Contemporary personality assessment relies heavily on psychometric scales, which offer efficiency but risk oversimplifying the rich and contextual nature of personality. Recognizing these limitations, this study explores the use of commercially available generative large language models (LLMs), such as ChatGPT, Claude and so on, to assess personality traits from open-ended qualitative narratives. Across two distinct samples and methodologies (spontaneous streams of thought and daily video diaries), we used seven commercial, generative LLMs to score Big-Five personality traits, achieving convergence with self-report measures comparable to or exceeding established benchmarks (for example, self-other agreement, ecological momentary assessment, and bespoke machine learning models). Although results differed across different LLMs, we found that using the average LLM score across models provided the strongest agreement with self-report. Further, LLM-generated trait scores also demonstrated predictive validity regarding daily behaviours and mental health outcomes. This LLM-based approach achieved quantitative rigour based on qualitative data and is easily accessible without specialized training. Importantly, our findings also reaffirm that personality is expressed ubiquitously, in that it is carried in the stream of our thoughts and is woven into the fabric of our daily lives. These results encourage broader adoption of generative LLMs for psychological assessment and-given the new generation of tools-stress the value of idiographic narratives as reliable sources of psychological insight.","PeriodicalId":19074,"journal":{"name":"Nature Human Behaviour","volume":"93 1","pages":""},"PeriodicalIF":29.9,"publicationDate":"2026-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146088911","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-28DOI: 10.1038/s41562-025-02381-5
Andrea I. Luppi, Lynn Uhrig, Jordy Tasserie, Pedro A. M. Mediano, Fernando E. Rosas, S. Parker Singleton, Daniel Gutierrez-Barragan, Silvia Gini, Pablo Castro, Camilo M. Signorelli, Daniel Golkowski, Andreas Ranft, Rüdiger Ilg, Denis Jordan, Kanako Muta, Junichi Hata, Hideyuki Okano, Zhen-Qi Liu, Yohan Yee, Alain Destexhe, Rodrigo Cofre, David K. Menon, Alessandro Gozzi, Bechir Jarraya, Emmanuel A. Stamatakis
The mammalian brain orchestrates the processing and integration of information to guide behaviour. Here, to characterize mammalian information-processing architecture, we combine functional neuroimaging and anaesthesia in humans, macaques, marmosets and mice. We show that breakdown of information integration is a convergent effect of diverse anaesthetics across mammalian species. As the system disintegrates, brain dynamics become more difficult to control. Both effects are reversed upon re-awakening induced by thalamic deep-brain stimulation in the macaque. Regional breakdown of integrated information coincides with the species-specific spatial topography of PVALB/Pvalb gene expression. To provide mechanistic insight beyond correlation, we develop computational models for humans, macaques and mice that integrate species-specific connectivity and transcriptomic gradients, demonstrating their respective roles for controlling brain dynamics and information integration. We reveal evolutionarily conserved controllers of information integration in the mammalian brain.
{"title":"Convergent transcriptomic and connectomic controllers of information integration and its anaesthetic breakdown across mammalian brains","authors":"Andrea I. Luppi, Lynn Uhrig, Jordy Tasserie, Pedro A. M. Mediano, Fernando E. Rosas, S. Parker Singleton, Daniel Gutierrez-Barragan, Silvia Gini, Pablo Castro, Camilo M. Signorelli, Daniel Golkowski, Andreas Ranft, Rüdiger Ilg, Denis Jordan, Kanako Muta, Junichi Hata, Hideyuki Okano, Zhen-Qi Liu, Yohan Yee, Alain Destexhe, Rodrigo Cofre, David K. Menon, Alessandro Gozzi, Bechir Jarraya, Emmanuel A. Stamatakis","doi":"10.1038/s41562-025-02381-5","DOIUrl":"https://doi.org/10.1038/s41562-025-02381-5","url":null,"abstract":"The mammalian brain orchestrates the processing and integration of information to guide behaviour. Here, to characterize mammalian information-processing architecture, we combine functional neuroimaging and anaesthesia in humans, macaques, marmosets and mice. We show that breakdown of information integration is a convergent effect of diverse anaesthetics across mammalian species. As the system disintegrates, brain dynamics become more difficult to control. Both effects are reversed upon re-awakening induced by thalamic deep-brain stimulation in the macaque. Regional breakdown of integrated information coincides with the species-specific spatial topography of PVALB/Pvalb gene expression. To provide mechanistic insight beyond correlation, we develop computational models for humans, macaques and mice that integrate species-specific connectivity and transcriptomic gradients, demonstrating their respective roles for controlling brain dynamics and information integration. We reveal evolutionarily conserved controllers of information integration in the mammalian brain.","PeriodicalId":19074,"journal":{"name":"Nature Human Behaviour","volume":"15 1","pages":""},"PeriodicalIF":29.9,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146057205","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"心理学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号